Deep Learning and Ensemble Learning for Traffic Load Prediction in Real Network

Abstract

For Internet Service Providers (ISPs), network traffic load prediction enables various practical applications such as load balancing, network planning, and network maintenance. With these applications, traffic load prediction is regarded as an important technology for developing intelligent network management and predictive maintenance. Predictive maintenance allows ISPs to be cost-effective. Traffic load prediction can assist humans in decision-making and increase automation. To predict traffic load, we apply deep-learning and ensemble-learning approaches. The main contributions of this paper are: (1) we formulate the network traffic load prediction problem, and implement a deep-learning-based system to resolve it; (2) we propose an ensemble-learning method that leverages multiple deep-learning models to obtain better predictive performance than any of the constituent deep-learning models; and (3) we evaluate the models using the real data.